Barrett's esophagus is a highly premalignant disease of unknown prevalence, but it predisposes to the development of esophageal adenocarcinoma, which is increasing at alarming rates in Western countries. The molecular genetics of esophageal adenocarcinoma and its precursor lesion, Barrett's esophagus, has been studied intensively in recent years. However, a better knowledge of the molecular alterations occurring in this setting will yield several benefits. Firstly, the discovery of novel molecular alterations will yield clues to biological pathways underlying Barrett's-associated neoplastic transformation, and these clues may lead to better in vitro and in vivo models of this disease. Secondly, molecular alterations themselves can be used as markers of early detection, disease progression, or ultimate prognosis in patients with Barrett's or cancer. Thirdly, these molecular alterations can be pursued as possible therapeutic targets for intervention, in both the prevention and treatment of this disease. The Aims of the current proposal will be to discover novel molecular alterations in Barrett's metaplasia and neoplasia, and to concentrate on the second of these benefits, i.e., to perform translational research to determine the potential value of these alterations as markers of disease progression. By using the same cDNA microarray platform to determine changes in DNA copy number, methylation status, and gene expression level, we will facilitate the translation of molecular genetic data from the genomic, to the epigenetic, to the transcriptomic, and finally to the protein (biomarker) level. This final level will employ tissue microarrays to test and validate specific candidate genes derived from the first three levels of study. These goals will be implemented by pursuing the following Aims: Aim 1. To perform global exploration for changes in DNA copy number in the Barrett's metaplasia-dysplasia-adenocarcinoma sequence (Barrett's neoplasia), using cDNA microarray-based comparative genomic hybridization (microarray-CGH). a) Global patterns of DNA amplification and deletion will be identified by microarray-CGH and characterized at each stage of Barrett's neoplasia, using hierarchical clustering, significance analysis of microarrays (SAM), and artificial neural networks, b) Specific cDNAs showing the most consistent and/or marked alterations in DNA copy number will be identified, characterized and validated using quantitative real-time PCR, for further study in Aim 4. Aim 2. To perform global epigenetic profiling at various stages in Barrett's neoplasia, using methylation-specific oligonucleotide microarrays. The genome will be screened for novel targets of DNA hypermethylation in various stages of Barrett's neoplasia, using methylation-specific oligonucleotide microarrays. Aim 3. To Using ANNs, to perform analyses of global gene expression data in Barrett's neoplasia. Results of genomic studies in Aims 1 and 2 will be correlated with global expression data in order to identify the genes most significantly different at both the genomic and transcriptomic levels at each stage of Barrett's neoplasia, for further study in Aim 4. Aim 4. Using tissue microarrays, to evaluate and validate potential biomarkers at the protein level in Barrett's neoplasia. Potential biomarkers identified in Aims 1-3 will be studied individually for expression at the protein level in all stages of Barrett's neoplasia, using tissue microarrays.